Build and Stress-Test a Multi-Asset Portfolio with Python
I'll be attending Build and Stress-Test a Multi-Asset Portfolio with Python, a live, hands-on online workshop with Cordell Tanny, CFA, FRM, FDP, on Saturday, November 28, 2026.
9:30 AM – 1:00 PM EDT / 3:30 PM – 7:00 PM CET — fully online.
Most portfolio courses stop at mean-variance optimization and call it a day. This one doesn't: four hours, entirely in Python, on a 12-asset universe spanning equities, government bonds, credit, inflation protection, real assets, and cash.
What caught my attention isn't the asset allocation part — it's what comes after:
- Replacing naive diversification with hierarchical clustering to recognize assets providing genuine diversification
- Modeling returns with a Johnson SU distribution and simulating them via block bootstrapping, instead of pretending markets are normally distributed
- Building CVaR-based and downside-target-based portfolios alongside mean-variance ones, then evaluating allocation shifts and downside risk via Sortino, Calmar, Omega, and Cornish-Fisher-adjusted Sharpe ratios
- Stress-testing across different market regimes, and desmoothing illiquid-asset returns to confront the stale pricing that understates their true risk
- Closing with an LLM-assisted capstone that turns the quantitative output of portfolio risk decomposition and stress-test results into a readable, defensible risk memo
Apart from this teaser, I'll be sharing a short series around this workshop — three posts on the ideas that stood out most (CVaR, hierarchical clustering, and the Johnson SU distribution), and a review after the workshop.
If quantitative portfolio construction is part of your work — asset allocation, risk management, or research — this is worth about four hours.
Full details and registration: https://luma.com/e0iytryu?coupon=OZIMEK40&utm_source=expert
Disclosure: I received a complimentary ticket to this event from Packt Publishing. The views above are my own.
